Skip to content

Latest commit

 

History

History
256 lines (190 loc) · 5.89 KB

API.md

File metadata and controls

256 lines (190 loc) · 5.89 KB
Raw

Startup Options

 $ norch --help
 
   Usage: norch [options]
 
   Options:
 
    -h, --help                   output usage information
    -V, --version                output the version number
    -p, --port <port>            specify the port, defaults to 3030
    -i, --norchHome <indexPath>  specify the name of the index directory, defaults to norch-index
    -l, --logLevel <logLevel>    specify the loglevel- silly | debug | verbose | info | warn | error
    -m, --machineReadable        machine readable, logo not printed, all stdout/stderr is JSON

Endpoints

Endpoint Method Response Typical Use Case
/add POST status code Add documents to the index
/availableFields GET stream Discover the name of fields which can be searched in
/buckets GET stream Aggregate documents on ranges of metadata
/categorize GET stream Aggregate documents on single metadata values
/concurrentAdd POST status code For when more than one source is adding documents to the index at the same time
/createSnapshot POST status code Create a snapshot of the index
/delete DELETE status code Remove documents from index
/docCount GET object Counts total document in index
/flush DELETE status code Remove all documents from index
/get GET stream Get documents by ID
/import POST file Import/merge an existing index into this one
/latestSnapshot GET file Download the latest index snapshot
/listSnapshots GET file See list of snapshots
/match GET stream Match by linguistic similarity- autosuggest, autocomplete
/search GET stream Search in the index
/totalHits GET object Show number of hits that a given query returns

API

GET

/availableFields

TODO

/buckets

Wrapper for search-index's buckets method. Get user defined aggregations of documents.

Query parameter: q

Example:

curl -X GET http://localhost:3030/buckets -G --data-urlencode q@- <<REQUEST_BODY
{
  query: [{
    AND: {
      '*': ['reuter']
    }
  }],
  buckets: [{
    field: 'topics',
    gte: 'barley',
    lte: 'barley',
    set: true
  }, {
    field: 'topics',
    gte: 'lin',
    lte: 'oat',
    set: true
  }]
}
REQUEST_BODY

/categorize

Wrapper for search-index's categorize method. Aggregate documents on metadata: Example: show counts by topic value

Query parameter: q

Example:

curl -X GET http://localhost:3030/categorize -G --data-urlencode q@- <<REQUEST_BODY
{
  "query": [{
    "AND": {
      "*": ["reuter"]
    }
  }],
  "category": {
    "field": "topics"
  }
}
REQUEST_BODY

/docCount

Count the documents in the index

/get

Get documents by ID

Query parameter: ids An array of document ids

Example:

curl -X GET http://localhost:3030/get -G --data-urlencode ids@- <<REQUEST_BODY
["3", "7"]
REQUEST_BODY

/latestSnapshot

Return the latest snapshot of the index. Generated the last time /snapshot was run

/listSnapshot

Show a list of all available snapshots

/match

Wrapper for search-index's match method. Returns word suggestions based on frequency in the index, used for making autosuggest and autocomplete functions.

Query parameter: q

Example:

curl -X GET http://localhost:3030/match -G --data-urlencode q@- <<REQUEST_BODY
{
  "beginsWith": "lon"
}
REQUEST_BODY

/search

Wrapper for search-index's search method. Search in the index. Filtering on buckets and categories are applied as fielded search for the regular query object.

Query parameter: q

Example:

curl -X GET http://localhost:3030/search -G --data-urlencode q@- <<REQUEST_BODY
{
  "query": {
    "AND": {
      "*": ["marathon"]
    }
  }
}
REQUEST_BODY

Example, filtering on buckets:

curl -X GET http://localhost:3030/search -G --data-urlencode q@- <<REQUEST_BODY
{
  "query": {
    "AND": {
      "*": ["marathon"],
      "finishingTime": [{
        "gte": "7200",
        "lte": "8999"
      }]
    }
  }
}
REQUEST_BODY

Example, filtering on categories:

curl -X GET http://localhost:3030/search -G --data-urlencode q@- <<REQUEST_BODY
{
  "query": {
    "AND": {
      "*": ["marathon"],
      "equipment": ["shoes"]
    }
  }
}
REQUEST_BODY

/totalHits

Show total number of hits for a qiven query. Queries take the same format as for /search

POST

/add

Add documents to the index.

Example: (where justTen.json is a newline separated object stream)

curl -X POST -d @justTen.str http://localhost:3030/add

API: https://github.com/fergiemcdowall/search-index/blob/master/docs/API.md#defaultpipeline

/concurrentAdd

Use this method when several sources are adding documents simultaneously

/createSnapshot

Creates a snapshot of the index which is then available for export under /latestSnapshot

/import

Import or merge an existing snapshot into this index

DELETE

/delete

Deletes documents from the index

Query parameter: ids An array of document ids

Example:

curl -X DELETE http://localhost:3030/delete -G --data-urlencode ids@- <<REQUEST_BODY
["7", "1"]      
REQUEST_BODY

/flush

Deletes all documents from index